The presence of large and many nevi (moles) is the most important predictor of melanoma occurrence. While nevi frequency increases with age, implicating sun exposure in their development, additive genetic factors appear to explain the majority of inter-individual variation in nevus size and counts. Data from adolescent twins studies in the UK and in Australia show that between 60 and 70% of nevus count variation is explained by genetic factors, although the genes responsible for this variation are unknown. Our previous work has demonstrated that in the absence of genetic variation, there appears to be no increased risk of melanoma associated with large or many nevi, implying that among individuals with large nevi, only the genetically susceptible will develop melanoma. Much work has been done on the occurrence of candidate genes in familial melanomas, but none have stood out that are both sufficiently penetrant and prevalent that they could contribute to a majority of melanomas. There is ample evidence of environmental causation in melanoma, by virtue of sun exposure. To further our understanding of the role of genotype in melanoma etiology, we are developing a study of gene-environment interaction in a population-based cohort of 52,000 twins in California, a state with among the highest rates of melanoma in the world. In the next few years, over 1,000 melanomas will have been diagnosed in this cohort, making it a critical resource for identifying gene-environment interactions in melanoma etiology, by use of traditional twin methods to partition additive genetic effects and those borne of shared environment. In this application we aim to conduct a feasibility study that will greatly enhance the success of the future study of melanoma etiology in the California Twin Program (CTP). Our specific aims are: 1. determine the feasibility of conducting a large scale twin study of multiple genetic and environmental determinants of melanoma: (a) select and contact identical twins with varying levels of melanoma risk genotype from the population-based California Twin Program cohort (b) gather, isolate, amplify and store samples of DNA from buccal swabs from each twin (c) demonstrate that the prevalence of key genes thought to confer melanoma risk is sufficient to justify a large-scale study 2. examine whether or not the prevalence of key genes conferring melanoma risk is higher in twin pairs with high melanoma risk genotype based on their nevi concordance than among those with low melanoma risk genotype 3. validate self-report of nevus counts (against physician assessment) and sun exposure (against repeated self-report) in this twin population 4. construct a theoretical model for the relationship between key genes conferring melanoma risk, the presence of large nevi, and sun exposure, that can be tested in a large scale twin study. This work will set the scene for future studies assessing the relative importance of genes and environment in melanoma risk.